Measuring Weight in a Neonatal Care System with a Sensing Array

The present disclosure generally relates to neonatal care systems and methods, and more particularly to systems and methods for a neonatal care system with integrated weighing.

Neonates, particularly premature infants, are often placed within neonatal care systems, such as an incubator and/or warmer, so that they may have a controlled and monitored environment to aid in their survival and growth. Accordingly, it is useful to monitor the infant's weight while the infant is in the incubator. It is additionally useful to monitor the infant’s weight because medical therapies, such as the dosing of pharmeceuticals, are based upon an accurate determination of the infant's weight. Accordingly, neonatal care systems such as incubators and warmers may include integrated weighing systems to determine a weight of an infant in the neonatal care system.

This Summary is provided to introduce a selection of concepts that are further described below in the Detailed Description. This Summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter.

A neonatal care system includes a processor and a memory. The memory includes instructions configured to cause the processor to receive strain amounts corresponding the strain sensors of a tactile sensor array that is disposed beneath a patient that is disposed on a mattress. Additionally, the instructions cause the processor to generate a strain map comprising a visualization of the strain amounts in association with the corresponding X, Y coordinate locations of the strain sensors. Further, the instructions cause the processor to generate a weight map based on the strain map and a strain-weight model. Additionally, the weight map includes weights corresponding to the strain amounts. Further, the instructions cause the processor to identify a patient location. Additionally, the instructions cause the processor to determine a weight of the patient based on the patient location and the weight map.

In one embodiment, the instructions are executable by the processor to determine that the weight map indicates a presence of an additional uniform weight on the strain sensor array, determine an amount of the additional uniform weight, and subtract the uniform weight from each of the weights.

In one embodiment, the instructions are executable by the processor to determine that the strain map indicates a presence of an additional gradient strain on the strain sensor array, determine gradient strain increases associated with a sloping subset of the strains, subtract the gradient strain increases from the sloping subset of strains, and generate the weight map based on a strain map resulting from subtracting the gradient strain increases.

In one embodiment, the strain sensors include piezoelectric tactile sensors.

In one embodiment, identifying the patient location includes identifying a contiguous set of strain values that correspond to a predetermined threshold that is indicative of a strain that the patient places on corresponding strain sensors.

In one embodiment, determining that the predetermined threshold is indicative of the strain includes determining a density of the patient, determining a weight value threshold for a sensor based on the density and a number of sensors on which the patient is located, and determining that each of weights corresponding to the contiguous set of strain values is less than or equal to the weight value threshold.

In one embodiment, determining the weight of the patient includes summing a subset of weights of the weight map that are associated with a subset of strain sensors corresponding to the patient location.

In one embodiment, identifying the patient location includes capturing an image of the patient on the mattress, identifying the patient within the image, overlaying the image on the strain map, and identifying a subset of strain sensors corresponding to where the patient and the subset of strain sensors overlap.

A method includes receiving weights corresponding to load cells of a load cell array that is disposed beneath a patient that is disposed on a mattress. Additionally, the method includes identifying a patient location. Further, the method includes generating a weight map having the weights. Additionally, the method includes determining a weight of the patient based on the patient location and the weight map.

In one embodiment, the method includes determining that the weight map indicates a presence of an additional uniform weight on the load cell array. Additionally, the method includes determining an amount of the additional uniform weight. Further, the method includes subtracting the uniform weight from each of the weights.

In one embodiment, identifying the patient location includes capturing an image of the patient on a mattress. Further, identifying the patient location includes identifying an image location of the patient within the image. Additionally, identifying the patient location includes overlaying the image on the weight map. Further, identifying the patient location includes identifying a subset of the load cells that overlap with the image location of the patient in the overlaid image.

In one embodiment, identifying the patient location includes receiving strain amounts corresponding to strain sensors in a strain sensor array. Additionally, identifying the patient location includes generating a strain map comprising strain amounts associated with corresponding X, Y coordinate locations for the strain sensors. Further, identifying the patient location includes identifying a patient location based on the strain map, and identifying a subset of the load cells that correspond to a subset of the strain sensors that overlap with the patient location.

In one embodiment, identifying the patient location comprises identifying a contiguous set of strain values in the strain map that correspond to a predetermined threshold that is indicative of a strain that the patient places on a corresponding plurality of strain sensors.

In one embodiment, determining the weight of the patient includes summing a subset of the weights associated with a subset of strain sensors corresponding to the patient location.

A computer-readable storage medium includes instructions executable by a processor to receive strain amounts corresponding to strain sensors in a strain sensor array that is disposed beneath a patient that is disposed on a mattress. Additionally, the instructions are executable by the processor to generate a strain map having strain amounts associated with corresponding X, Y coordinate locations for the strain sensors. Further, the instructions are executable by the processor to generate a weight map based on the strain map and a strain-weight model, the weight map having weights corresponding to the strain amounts. Additionally, the instructions are executable by the processor to identify a patient location. Further, the instructions are executable by the processor to determine a weight of the patient based on the patient location and the weight map.

In one embodiment, the instructions are executable by the processor to determine that the weight map indicates a presence of an additional uniform weight on the strain sensor array, determine an amount of the additional uniform weight, and subtract the uniform weight from each of the weights.

In one embodiment, the instructions are executable by the processor to determine that the strain map indicates a presence of an additional gradient strain on the strain sensor array, determine gradient strain increases associated with a sloping subset of the strains, subtract the gradient strain increases from the sloping subset of strains, and generate the weight map based on a strain map resulting from subtracting the gradient strain increases.

In one embodiment, the strain sensors include piezoelectric tactile sensors.

In one embodiment, identifying the patient location includes identifying a contiguous set of strain values that correspond to a predetermined threshold that is indicative of a strain that the patient places on corresponding strain sensors.

Various other features, objects, and advantages of the invention will be made apparent from the following description taken together with the drawings.

In the present description, certain terms have been used for brevity, clarity and understanding. No unnecessary limitations are to be inferred therefrom beyond the requirement of the prior art because such terms are used for descriptive purposes only and are intended to be broadly construed.

As used herein, unless otherwise limited or defined, discussion of particular directions is provided by example only, with regard to particular embodiments or relevant illustrations. For example, discussion of "top," "bottom," "front," "rear," "left," "right," "horizontal," "vertical," and "longitudinal" features and/or relative motion, e.g., movement "up" and "down," is generally intended as a description only of the orientation of such features relative to a reference frame of a particular example or illustration. Correspondingly, for example, a "top" feature may sometimes be disposed below a "bottom" feature (and so on), in some arrangements or embodiments. Additionally, or alternatively, embodiments may be arranged in a different orientation such that "top" and "bottom" features are arranged horizontally relative to each other, for example in a "left-to-right" orientation.

The use herein of the terms "including," "comprising," or "having," and variations thereof, is meant to encompass the elements listed thereafter and equivalents thereof, as well as additional elements. Embodiments recited as "including," "comprising," or "having" certain elements are also contemplated as "consisting essentially of" and "consisting of" those certain elements.

The inventors have recognized a problem with current neonatal care systems, such as incubators, infant warmers, and other types of neonatal care systems and devices. As stated previously, neonatal care systems such as incubators and warmers may include integrated weighing systems to determine the weight of an infant in the system. Typically, a neonate’s weight may be measured once a day to monitor growth and to determine the appropriate medication dosages and/or intravenous (IV) fluid intake. Currently, a healthcare provider may lift the neonate (with any attached equipment) off the mattress while ensuring that arms, legs, blankets, and clothing do not touch the mattress and affect the weight measurement. This lifting helps in baselining (i.e., taring) the measure of the scale. Once the scale is baselined, the healthcare provider may place the neonate back onto the mattress while all other equipment is held off. Hence, the scale may provide a measure of the neonate’s weight. However, this process can interfere with the neonate’s neurodevelopment as the lifting and other manipulation is a negative stimulus, can interfere with neonate’s sleep, cause the neonate discomfort, and potentially cause a dangerous dislodgement of therapeutic tubes and/or sensors. Additionally, this lifting and measuring process may be somewhat cumbersome for the healthcare provider. Further, in contrast to a daily measurement, a more continuous measurement (e.g., once an hour) of the neonate’s weight can more closely track the neonate’s intake and output, and make it possible to determine an estimate of the neonate’s metabolism. However, a continuous measurement may be more harmful and labor-intensive than daily measurement because of the lifting process involved with current systems.

In view of the foregoing problems and challenges recognized by the inventors through their extensive research and experience in the field of neonatal care systems, the inventors have developed the disclosed improved systems and methods for weighing an infant housed in an incubator, warmer, or other neonatal care system. More specifically, some embodiments of the present disclosure may use a tactile sensor array embedded in the mattress to enable the continuous measurement of the neonate’s weight without disruption to the neonate’s neurodevelopment, sleep, and comfort. Additionally, such embodiments may provide a strain map and/or a camera to isolate and/or distinguish the neonate’s weight from the weight of other objects located on the mattress. In such embodiments, it may be possible to weigh the neonate without manual lifting, thus eliminating a negative stimulus to the neonate, a potential cause of tube dislodgement, and the cumbersome process of baselining and weighing. In these ways, such embodiments can provide continuous neonate weighing to enable quantification of inputs such as medication and intravenous fluids, and the quantification of outputs, e.g., waste. Further, such embodiments may determine the neonate’s weight distribution, e.g., the weight of the neonate’s head, versus the weight of the rest of the body, and the like. Additionally, such embodiments may make it possible to determine the metabolism of the neonate.

1 FIG. 10 10 8 is a perspective view of an exemplary neonatal care systemthat measures weight with a sensing array according to one embodiment of the present disclosure. The neonatal care systemis shown within a room, such as a labor and delivery suite, or a neonatal intensive care unit, within a medical facility. The ambient air temperature within the room is controlled by room thermostat, which is adjustable up and down according to the specification of the patient and medical personnel in a customary manner.

10 12 14 16 18 26 28 24 1 24 26 32 34 1 34 12 50 2 FIG. The neonatal care systemmay define an infant warmer including a standsupported by legsand feetprovided with wheelsin a manner presently known in the art. The walls, and in the case of an incubator, a cover(See), generally surround and cover the mattress, to prevent the patientfrom falling from the mattressand also to maintain a controlled environment within the interior. The air within the interior defined by the walls(and when present, the cover) is also referred to as inside air. The heater, e.g., radiative heater, may be a heat generating device such as those used within the exemplary warmers described above. The patientis warmed using the heater. Additionally, the standalso supports an enclosure(e.g., drawer for storage).

20 12 20 70 10 22 12 20 22 38 24 1 38 38 38 24 70 38 38 38 38 24 24 A columnextends upwardly from the stand. The columnmay include a controller(e.g., a microprocessor, computer processing circuit, and the like) for operating the neonatal care systemin a manner presently known in the art. The platformmay be supported by the base on the stand, and may be height adjustable along the columnin a manner presently known in the art. Further, the platformis configured to support a sensing array, which may be incorporated (e.g., embedded) within and/or below the mattress, which is configured to support the patient. The sensing arraymay include a tactile sensor array on a flexible base, such as a piezoresistive, piezoelectric, or a capacitive array, used as a touch sensor for tactile perception. In response to a tactile stimulus, the sensing arraymay provide signals indicating information about forces at the points of contact. Further the sensing arraymay be a flexible, stretchable, and thin material. According to some embodiments of the present disclosure, the array is a piezoresistive array that may be embedded within the mattress. Additionally, the controllermay include a scale manager, or a portion thereof, for performing weighing with the sensing arrayas described herein. The sensing arraycan measure a strain imposed on the array by anything and/or anyone located thereon. Additionally, the sensing arraymay include a load cell array that is located below the sensing array. In some embodiments of the present disclosure, the load cell array may be embedded within the mattress. Alternatively, the load cell array may be located beneath the mattress. Each load cell of the load cell array may be located beneath one or more elements of the tactile sensor array. Further, the load cell array may be a set of force transducers, each of which may produce a signal that is indicative of the weight located thereon. Although the term, load cell, is used in the disclosure, it should be understood that the load cell array may incorporate any sensor(s) or device(s) that generates a signal that can be measured, and is representative of weight or force on each element of the load cell array.

38 1 1 24 The scale manager may periodically use the sensing arrayto take a weight measurement of the patientwithout lifting the patientfrom the mattress. More specifically, the scale manager may identify a set of sensing array elements on which the patient’s body is located. Additionally, the scale manager may determine the patient’s weight by correlating the strain on each sensing array element to a measured weight, and summing the measured weights correlating to each sensing array element on which the patient’s body is located. Alternatively, the scale manager may identify the load cells located beneath the sensing array elements on which the patient’s body is located. In such a scenario, the scale manager may determine the patient weight by summing the weights measured by the identified load cells. In embodiments where the load cell array is located beneath the mattress, the mattress weight may show up as an increased weight value that is uniform across all the load cells. Accordingly, some embodiments of the present disclosure may subtract this uniform value from the measured weight.

1 24 38 10 48 48 1 24 1 24 38 As stated previously, the scale manager may take periodic weight measurements. In this way, the scale manager may generate a data feed of weight measurements. These periods can be one or more times a second, several seconds, one or more minutes, one or more hours, and the like. As such, it is possible that a healthcare provider, or other person, may place objects on (or remove objects from) the patientand/or mattress, which may affect the measured weight. However, in order to determine the weight measurement more accurately, it may be useful to identify such objects and exclude the weight of these objects from the weight measurement. Accordingly, in one embodiment of the present disclosure, the scale manager may identify the location of such objects based on the output of the sensing array. Additionally, or alternatively, the neonatal care systemmay include an image sensor. The image sensormay capture individual images, and/or video, of the patientand mattress. Further, the scale manager may analyze these images to identify where the patientis located on the mattress, and thus, identify the elements of the sensing arrayand/or load cell array that are useful for measuring the patient’s weight.

10 40 42 10 40 44 46 44 46 10 The neonatal care systemfurther includes a user interface, which may include a displayconfigured to provide warning indications (text, colors, icons, and the like) as well as messages relating to operation of the neonatal care system. Additionally, the user interfacemay include a speakerand one or more lights. The speakerand lightsmay provide further information regarding the operational status of the neonatal care system.

44 46 40 10 40 10 42 44 46 42 44 46 10 10 Additionally, the speakerand lightsmay communicate information to a healthcare provider and/or operator via sounds, spoken text, spoken words, flashing, varying colors, and/or the lights being on or off. In this manner, as is discussed further below, the user interfaceprovides feedback customary of infant care systemspresently known in the art, but also additional information, warnings, and/or the like according to the present disclosure. It should be recognized that the user interfacemay also or alternatively be provided via an external device (e.g., a mobile device such as a tablet or smart phone) in communication with the neonatal care system. For example, a smart phone may serve as the display, speaker, and/or lights(alone or in conjunction with another display, speaker, and lights, on the neonatal care system) that communicates with the neonatal care systemvia Bluetooth® or another wireless protocol known in the art.

2 FIG. 1 FIG. 1 FIG. 2 FIG. 2 FIG. 2 FIG. 2 FIG. 1 FIG. 1 FIG. 10 38 10 10 12 22 24 26 34 38 40 48 50 70 1 34 24 22 10 28 10 26 28 30 26 28 1 24 22 26 28 48 70 48 70 70 38 is a perspective view of an exemplary neonatal care systemthat measures weight with a sensing arrayaccording to one embodiment of the present disclosure. In this example, the neonatal care systemis similar to that of, but as an incubator rather than an infant warmer. Similar to, the neonatal care systemofincludes stand, platform, mattress, walls, heater, sensing array, user interface, image sensor, enclosure, and controller. In the incubator, the patientis warmed using warm air flowing in the incubator from the heaterand a fan (not shown) located below the mattressand platform. Additionally, the neonatal care systemofincludes a cover, whereby the interior of the neonatal care systemis defined by the wallsand the cover. Further, the incubator ofincludes portholeswithin the wallsand/or coverto provide access to the interior (e.g., patient, mattress, and/or the platform) without opening one or more of the wallsand/or the coverin a manner presently known in the art. The image sensorand controllerofmay be similar to the image sensorand controllerof. As such, the controllermay include a scale manager that may perform weighing with the sensing arrayand/or load cell array as described with respect to.

3 FIG.A 302 1 304 24 302 38 24 1 304 302 302 1 304 is a side view of an example sensing array, patient, extraneous object, and mattress, according to one embodiment of the present disclosure. The sensing arraymay be similar to the sensing array. In this example, the mattressis deformed by the weight of the patientand extraneous objectlocated thereon. Accordingly, the sensing arraymay also deform with the mattress. However, the sensors of the sensing arraymay or may not deform. Rather, the sensors respond to the forces and/or pressure of the patientand extraneous object. For example, piezoelectric sensors may generate a voltage in response to the forces and/or pressure.

3 FIG.B 4 4 FIGS.A andB 302 1 304 1 304 2 24 302 306 306 304 1 24 306 306 1 306 2 306 3 is a top view of the example sensing array, patient, extraneous objects-,-, and mattress, according to one embodiment of the present disclosure. The example sensing arraymay include multiple piezoelectric sensorsarranged in a grid pattern. According to some embodiments of the present disclosure, each of the piezoelectric sensorsmay produce an electric current in response to the amount of strain due to the weight of the portion of the extraneous objectand patientthat falls on that sensor location on the mattress. In this example, the pattern of the sensors’ outline represents the amount of force applied to the respective sensor. More specifically, the sensors-represent no force; the sensors-represent a lower amount of force; and, the sensors-represent a larger amount of force. For the purpose of clarity, in this figure, the terms lower and larger are used to describe the force here. However, the amount of force is described in greater detail with respect to.

4 FIG.A 400 306 302 400 302 400 306 24 400 3 306 302 306 400 400 306 304 1 1 304 2 304 1 304 2 is an example strain mapA due to the amount of force applied to each sensorin the sensing array, according to one embodiment of the present disclosure. According to some embodiments of the present disclosure, the scale manager may generate the strain mapA based on the outputs of the sensing array. The strain mapA may be a graphical representation of the strain of the sensorswhich depends on the force applied by the elements located on the mattress. In this example, the strain mapA is represented as a three-dimensional (D) graph, where the X and Y dimensions represent the locations of the sensorsin the X and Y dimensions of the sensing array. Additionally, the Z dimension represents the amount of strain of the sensorat each X, Y location. For clarity, this example strain mapA represents strain generically, without reference to specific units. Further, in this example, the lines of the strain mapA represent the slopes between sensorshaving different values of strain. As shown, the extraneous object-appears to give rise to the relatively greatest levels of strain, the patientcreates relatively lower levels of strain, and extraneous object-cerates the relatively lowest levels of strain. In this way, the strain map 400A may indicate the presence of the three elements on the mattress 24: the patient 1 and extraneous objects-,-.

4 FIG.B 400 400 400 306 306 302 24 400 400 400 304 1 304 2 is an example strain mapB representative of the strain indicated in strain mapA, according to one embodiment of the present disclosure. In this example, the strain mapB is represented as a two-dimensional (2D) mapping, wherein each different level of strain of the sensorsis represented graphically at an X, Y location that corresponds to the X, Y location of the sensorin the sensing array. Further, the graphic representation indicates greater levels of strain with greater levels of hashing at each location. Hence, the locations with no hatching represent no strain, and the locations with the greatest number of hatch lines represent the greatest relative strain. According to some embodiments of the present disclosure, the scale manager may determine outlines of elements located on the mattressbased on the strain mapA orB. Accordingly, the strain mapB includes outlines for each of the patient 1, and extraneous objects-,-.

4 FIG.C 400 400 306 306 302 400 400 306 24 400 400 400 1 304 1 304 2 1 1 1 is an example weight mapC, according to one embodiment of the present disclosure. In this example, the weight mapC is represented as a two-dimensional (2D) mapping, wherein the weight value corresponding to the strain level at each of the sensorsis represented numerically at an X, Y location that corresponds to the X, Y location of the sensorin the sensing array. The numeric value may indicate a weight in grams, ounces, or other units of weight measure depending on the embodiment implemented. According to some embodiments of the present disclosure, the scale manager may generate the weight map 400C based on strain mapA or strain mapB. More specifically, the scale manager may determine the weight based on a strain-weight model that correlates the strain at a sensorwith a weight value. As stated previously, the scale manager may determine outlines of elements located on the mattressbased on the strain mapA and/orB. Accordingly, the strain mapB includes outlines for each of the patient, and extraneous objects-,-. Hence, the scale manager may determine the patient weight may by adding the weights indicated at the sensors within the outline for the patient. In these ways, some embodiments of the present disclosure may determine the weight of the patienton a continuous basis (e.g., every hour) without lifting the patient, and thus avoiding the negative effects of lifting on neurodevelopment, sleep, and patient comfort.

5 FIG. 500 500 1 is a flow diagram depicting an exemplary processfor measuring weight in a neonatal care system with a sensing array according to one embodiment of the present disclosure. The scale manager may perform the processto measure the weight of the patient on a continuous basis, and without disturbing the patient.

502 302 302 24 10 306 306 400 306 At operation, the scale manager may generate a strain map based on the output of the sensing array. As stated previously, the sensing arraymay include a piezoresistive tactile sensor array that is embedded within the mattressof a neonatal care system. Generating the strain map may involve generating a graphical representation of strain. This graphical representation may include a series of sloping lines between neighboring points in 3D space. The X, Y locations of these points may represent the X, Y locations of the sensorsof the sensing array. Thus, the Z coordinate may represent the amount of strain indicated by the electric current generated by the sensorat the X, Y location. Alternatively, generating the strain map (e.g., strain mapB) may involve populating a two-dimensional array with numeric strain values, where the location of each element of the array corresponds to the X, Y location of the sensorhaving the corresponding strain value.

504 306 306 306 306 400 304 1 304 1 400 304 1 1 304 2 4 FIG.B At operation, the scale manager may generate the object outlines based on the strain levels and strain slope. More specifically, the scale manager may determine outlines of the elements on the mattress by identifying contiguous sets of sensorsfor each of the objects based on strain levels and strain slope. According to some embodiments of the present disclosure, the scale manager may use the sensorsassociated with each of the high strain points as the starting points for each object. The high strain points may be the locations in the sensing array where the sensorssense the relatively highest strain values within a contiguous set of sensors. For example, in the example strain mapB, described with respect to, the highest strain points are the locations within the extraneous object-, indicated by the heaviest hatching. Hence, the scale manager may identify the extraneous object-by determining the contiguous points with the high strain point. Further, the high strain point for the patient is also indicated by the locations in the example strain mapB with the relatively heaviest hatching. Accordingly, the scale manager may identify these locations as high strain points. Similar to identifying the extraneous object-, the scale manager may identify the contiguous points with the patient’s high strain point, and thus identify the outline of the patient. The scale manager may similarly identify the outline of extraneous object-. Accordingly, the scale manager may determine the weight of each using the strains indicated in the corresponding locations of the strain map and the strain-weight model.

506 1 304 1 304 2 1 304 1 304 2 1 1 304 1 304 2 1 304 1 304 2 1 1 1 1 306 1 1 306 306 306 1 1 1 306 306 At operation, the scale manager may identify the patientfrom the identified objects. According to some embodiments of the present disclosure, the scale manager may identify the patient based on the strain levels (e.g., weight), outline size (e.g., area), and strain slope. For example, objects-and-occupy a relatively small area, e.g., below a predetermined threshold for the area the patientoccupies, and thus, the scale manager may determine that the extraneous objects-,-, are not the patient. For example, the scale manager may calculate the predetermined threshold for the area occupied by the patientbased on a measurement of the patient’s length and width, potentially measured by a healthcare provider. Alternatively, or additionally, the scale manager may compare the weight of the extraneous objects-,-to a threshold weight (e.g., a previous weight measurement) for the patient, and determine that the extraneous objects-,-, are not the patient. For example, the scale manager may determine that the weight of any object that exceeds a predetermined threshold change from a previous measurement, is too light (or heavy) to be the patient, and thus, is not the patient. Alternatively, the scale manager may determine a density of the patient based on the previous weight and size measurements. Accordingly, the scale manager may determine a high weight threshold from the patientthat can be applied to each sensor. Hence, the scale manager may exclude any object with a weight exceeding this density from being considered as being the patient. Additionally, or alternatively, the scale manager may compare the strain slope of each identified object to a predetermined strain slope threshold to identify the patient. More specifically, the relative strain sensed by each of the sensorsunder the patient may not exceed the predetermined threshold strain slope. The strain slope refers to the difference in detected strain between contiguous sensors. For example, the weight of a dense object may merely fall on the sensorson which the dense object is located. Thus, the neighboring sensors may detect no strain. In such a case, the dense object may have a high strain slope (potentially infinite) at the edges, and a relatively low (potentially zero) strain slope elsewhere within the object outline. Accordingly, to determine the threshold strain slope, the scale manager may use the high weight threshold to determine a high strain slope threshold at the outline of the patient. Thus, any strain slope exceeding the threshold strain slope may not indicate the location of the patient. In these ways, the scale manager may determine that an identified object is not the patientbecause the strain slope between the set of sensorsassociated with the object exceeds a predetermined threshold, the strain exceeds a predetermined threshold, the weight exceeds a predetermined threshold, and/or the size of the object (indicated by the set of identified sensors) is less than a predetermined threshold.

508 400 400 306 302 306 24 At operation, the scale manager may generate a weight map, based on the strain map, e.g., strain mapA,B, and a strain-weight model. As stated previously, the strain map may indicate the amount of strain of (e.g., due to force applied) at each sensorof the sensing array. Further, the strain-weight model may correlate an amount of strain with a weight. More specifically, the strain-weight model may be a strain-weight calibration that is specific to the sensors. Alternatively, the scale manager, or another process, may generate the strain-weight model experimentally for the use scenario. For example, an operator may place calibrated weights on the mattress, and configure the scale manager to correlate the detected strain to the calibrated weight. In another alternative, the strain-weight model may be a look-up table or a functional equation.

306 306 302 306 1 Accordingly, the scale manager may generate a 2D table where each of the sensorsis represented at an X, Y location that corresponds to the X, Y location of the sensorin the sensing array. Further, the scale manager may determine the strain from each of the sensors, and populate each of the table locations with a numerical weight based on the weight corresponding to the strain according to the strain-weight model. According to some embodiments of the present disclosure, the scale manager may generate the weight map before identifying the patient.

510 1 306 At operation, the scale manager may determine the weight of the patient. Determining the weight of the patient may involve summing the weight corresponding to all the sensorsthat the scale manager identifies as being strained by the patient’s weight.

6 FIG. 600 600 1 is a flow diagram depicting an exemplary processfor measuring weight in a neonatal care system with a sensing array according to one embodiment of the present disclosure. The scale manager may perform the processto measure the weight of the patient on a continuous basis, and without disturbing the patient.

602 302 302 302 306 At operation, the scale manager may generate a strain map based on the output of the sensing array. As stated previously, the sensing arraymay include a piezoresistive tactile sensor array that is embedded within the mattress of a neonatal care system. Accordingly, the scale manager may use the sensor values from the sensing arrayto populate a two-dimensional table where each element in the table includes the sensor value of the sensorat a location corresponding to the location in the table.

604 400 48 1 304 24 400 At operation, the scale manager may overlay a captured image on a strain map (e.g., strain mapA). As stated previously, the image sensormay capture an image of the patient, extraneous objects, and the mattress. Thus, overlaying the captured image on the strain mapA may involve correlating the orientation of the strain map with the orientation of the captured image.

606 1 400 1 1 306 At operation, the scale manager may identify the patienton the strain mapA using the overlaid image. Identifying the patientmay involve identifying the sensors 306 on which the patientis located in the overlaid image. According to some embodiments of the present disclosure, identifying the patient may involve the use of a machine learning model trained to identify images of infants. Thus, the scale manager may use a recurrent neural network to identify the body of the patient in the image. However, other machine learning models may also, or alternatively, be used, such as, convolutional neural network (CNN), you only look once (YOLO), support vector machines (SVM), and the like. Further, the scale manager may identify the sensors 306 beneath the patient by using the image overlay to determine which of the sensorsoverlap with the portions of the image where the patient is located.

608 400 400 400 306 302 306 306 302 306 At operation, the scale manager may generate a weight map based on the strain map, e.g., strain mapA/B and a strain-weight model. As stated previously, the strain mapA may indicate the amount of strain of (e.g., due to force applied on) each sensorof the sensing array. Further, the strain-weight model may correlate an amount of strain with a weight. Accordingly, the scale manager may generate a 2D table where each of the sensorsis represented at an X, Y location that corresponds to the X, Y location of the sensorin the sensing array. Further, the scale manager may determine the strain from each of the sensors, and populate each of the table locations with a numerical weight based on the weight corresponding to the strain according to the strain-weight model.

610 1 1 306 1 At operation, the scale manager may determine the weight of the patient. Determining the weight of the patientmay involve identifying the weights in the weight map associated with sensorsidentified in the overlaid image. Additionally, determining the weight of the patientmay involve summing the identified weights.

7 FIG.A 702 1 704 708 24 702 38 708 710 710 702 708 24 is a side view of an example sensing array, patient, extraneous object, load cell array, and mattress, according to one embodiment of the present disclosure. The sensing arraymay be similar to the sensing array. Further, the load cell arraymay include multiple load cells, where a load cellmay be positioned beneath each sensor of the sensing array. In this example, the load cell arrayis located beneath the mattress.

7 FIG.B 702 1 704 1 704 2 24 702 706 706 704 1 24 708 706 708 706 708 706 708 708 24 708 24 is a top view of the example sensing array, patient, extraneous objects-,-, and mattress, according to one embodiment of the present disclosure. The example sensing arraymay include multiple piezoelectric sensorsarranged in a grid pattern. According to some embodiments of the present disclosure, each of the piezoelectric sensorsmay produce an electric current in response to the amount of force applied from the weight of the portion of the extraneous objectand patientlocated on the mattress. Additionally, a load cell(not shown) may be positioned below each of the sensors. Alternatively, there may be fewer load cellsthan strain sensors. As such, each load cellmay be positioned beneath multiple strain sensors. Accordingly, each load cellmay measure weight based on the strain sensed by the load cell’s corresponding set of sensors. Further, while the load cellsin this example are located beneath the mattress, as stated previously, the load cellsmay, alternatively, be embedded within the mattress.

7 FIG.C 4 FIG.B 5 6 FIGS., 700 706 702 700 400 706 700 706 704 1 704 2 is an example strain mapC illustrating the strain indicated at each sensorin the sensing arrayaccording to one embodiment of the present disclosure. The strain mapC may be similar to the strain mapB described with respect to. As stated previously, the scale manager may identify the sensorspositioned beneath the patient, as described with respect to. Accordingly, the strain mapC includes outlines around the strain indicators for the sensorsunder each of the patient 1 and extraneous objects-,-.

7 FIG.D 4 FIG.C 5 6 FIGS., 700 706 702 700 400 400 700 708 706 700 700 708 1 704 1 704 2 708 1 708 1 708 2 704 1 704 2 is an example weight mapD illustrating the weight indicated at each sensorin the sensing arrayaccording to one embodiment of the present disclosure. The weight mapD may be similar to the weight mapC, described with respect to. However, in contrast to the generation of the weight mapC using a strain-weight model, the scale manager may generate the weight mapD by using the weights measured by the load cellsunder the sensors. Similar to the strain mapC, the weight mapD includes outlines around the weight values for the load cellsunder each of the patientand extraneous objects-,-. Accordingly, as described with respect to, the scale manager may use the outlines to identify the load cellslocated beneath the patient. Thus, according to some embodiments of the present disclosure, the scale manager may determine the weight of the patient by summing the weights from the load cells-located under the patient, and disregard the weights from the load cells-located under the extraneous objects-,-.

8 FIG. 800 800 1 is a flow diagram depicting an exemplary processfor measuring weight in a neonatal care system with a sensing array according to one embodiment of the present disclosure. The scale manager may perform the processto measure the weight of the patient on a continuous basis, and without disturbing the patient.

802 708 708 24 1 704 1 704 2 7 7 FIGS.A throughD At operation, the scale manager may receive weight from a load cell array, such as the load cells, described with respect to. Each of the load cellsin the load cell array may measure a portion of the force on the mattress, including force caused by the weight of the patientand extraneous objects (e.g., extraneous objects-,-).

804 708 700 708 708 24 At operation, the scale manager may generate a weight map based on the received weights from the load cells. According to some embodiments of the present disclosure, the weight map (e.g., weight mapD) may be a two-dimensional array that includes a numeric value for the weight measured by each of the load cells. The two-dimensional array may correlate the weight measures with the location of the load cellswith respect to the mattress.

806 306 1 1 24 1 708 708 708 5 6 FIGS.and At operation, the scale manager may identify the patient location. Identifying the location of the patient may involve identifying which load cells can provide measures of the patient’s weight. Similar to identifying the sensorson which the patientis located (as described with respect to), the scale manager may identify the load cells on which the patient is located by capturing an image of the patienton the mattress, identifying the patientwithin the image, and identifying the load cellsthat overlap with the location of the patient in an overlaid image. According to some embodiments of the present disclosure, identifying the patient may involve the use of a recurrent neural network trained to identify images of infants. Thus, the scale manager may use the recurrent neural network to identify the body of the patient in the image. Further, the scale manager may identify the load cellsbeneath the patient by using the image overlay to determine which of the load cellsoverlap with the portions of the image where the patient is located.

702 700 702 706 706 700 708 706 708 706 708 706 708 1 706 Alternatively, the scale manager may identify the patient location using an image overlay with the sensing array. More specifically, the scale manager may generate a strain map (e.g., strain mapC) based on the strain values provided by the sensing array. Additionally, the scale manager may identify the sensorson which the patient is located using the captured image described above, to identify where the patient location overlaps with the sensorsin the image overlaid on the strain mapC. Further, the scale manager may identify which of the load cellscorrespond to the identified sensors. As stated previously, the load cellsmay be positioned beneath each sensor. Alternatively, there may be fewer load cellsthan sensors. As such, each load cellmay way the portion of the patientcorresponding to the location of multiple sensors.

808 700 At operation, the scale manager may calculate the patient’s weight based on the weight mapD and the patient location. More specifically, the scale manager may sum the measured weights from the load cells 708 that are identified as corresponding to the patient’s location as described above.

9 FIG.A 900 900 900 306 24 24 302 900 900 is an example strain mapA indicating the strain for each sensor location in a neonatal care scale weighing system with a sensing array according to one embodiment of the present disclosure. In this example, the strain mapA indicates the presence of an additional uniform strain on the sensing array. More specifically, there no zero strain values in the strain mapA. Rather, all of the sensorsindicate a level of strain, which may indicate that an extraneous object is located on the mattress, adding a uniform strain value that the sensors detect. For example, if a healthcare professional places a second mattress on the mattress, the second mattress may add a uniform strain detected by the sensor array. Thus, according to some embodiments of the present disclosure, the scale manager may determine that the sensing arrayis detecting a uniform strain by determining that the strain mapA has no zero values. Alternatively, the scale manager may use a threshold strain value to determine uniform strain. More specifically, if there are no strain values lesser than the threshold strain value, the scale manager may determine there is a uniform strain. Further, according to some embodiments of the present disclosure, the scale manager may plot a histogram of sensed strain values. The plotted histogram may indicate a minimum strain value. Hence, if the strain mapA has no strain values lesser thana the minimum value, the scale manager may determine there is a uniform strain.

9 FIG.B 900 900 900 900 is an example strain mapB indicating a modified strain for each sensor location in a neonatal care scale weighing system with a sensing array according to one embodiment of the present disclosure. In this example, the strain mapB represents the result of the scale manager modifying the strain mapA by removing (e.g., subtracting) the uniform strain from all strain values. In some embodiments, the scale manager may determine a uniform weight corresponding to the uniform strain using the strain-weight model. Accordingly, the scale manager may subtract the uniform weight from the weight map for the strain mapA.

10 FIG.A 1000 1000 1000 1000 is an example strain mapA indicating the strain for each sensor location in a neonatal care scale weighing system with a sensing array according to one embodiment of the present disclosure. In this example, the strain mapA represents a scenario where a healthcare provider has added a wedge to the mattress to prop up the patient’s head. Accordingly, the strain map indicates a presence of an additional gradient strain on the strain sensor array. In other words, the strain mapA indicates a sloping increase in strain values from right to left. Accordingly, the scale manager may identify the sensors showing the gradient strain increases, and subtract the gradient strain increases from the sloping subset of strains. According to some embodiments, the scale manager may identify the sloping increase or decrease and make a correcting modification using standard slope detection and correction methods. In one example, the scale manager may average the strain detected in each column (or row) of the sensing array, and identify a sloping increase based on continuous change in average slope across columns (or rows). Alternatively, the scale manager may identify an additional gradient weight in the weight map generated from the strain mapA. In such embodiments, the scale manager may identify the sloping increase or decrease in weight values, and make a correcting modification using standard slope detection and correction methods.

10 FIG.B 1000 1000 is an example strain mapB indicating a modified strain for each sensor location in a neonatal care scale weighing system with a sensing array according to one embodiment of the present disclosure. In this example, the scale manager has modified the strain mapA by subtracting the sloping increase in strain amounts.

11 FIG. 1 2 3 3 4 4 4 5 6 7 7 7 7 8 9 9 10 10 11 FIGS.,,A,B,A,B,C,,,A,B,C,D,,A,B,A,B, and 1100 1100 1100 1102 1104 1110 1112 1114 1102 1106 1104 1114 1102 1104 1110 1112 1114 is an exemplary scale managerfor measuring weight in a neonatal care system with a sensing array according to one embodiment of the present disclosure. The example scale managermay measure weight with a sensing array, as described with respect to. In this example, the scale managerincludes a processor, memory, input-output (I/O) interface, and network interface, which may be connected by an interconnect. The processormay be a computer processing circuit (e.g., a central processing unit (CPU)) that retrieves and executes programming instructionsstored in the memoryto perform the functionality described herein. The interconnectmay move data, such as programming instructions, between the processor, memory, I/O interface, and network interface. The interconnectmay include one or more buses.

1104 1104 1104 1106 1 2 3 3 4 4 4 5 6 7 7 7 7 8 9 9 10 10 11 FIGS.,,A,B,A,B,C,,,A,B,C,D,,A,B,A,B, and The memorymay be a computer memory or storage device, including volatile memory, such as a random access memory (RAM) device (e.g., static RAM, dynamic RAM, and the like), non-volatile memory, such as a hard disk drive, solid state device (SSD), removable memory cards, optical storage, flash memory devices, and the like. In some examples, the memorymay include volatile and non-volatile memory devices. Further, the memorymay store instructionsfor measuring weight with a sensing array as described with respect to.

1100 1116 1110 1118 1112 1116 38 48 708 42 44 46 1118 1100 1118 1 2 FIGS.and 7 7 7 7 FIGS.A,B,C, andD 1 2 FIGS.and Additionally, the scale managermay be in electronic communication with I/O devicesthrough the I/O interface, and with a networkthrough the network interface. The I/O devicesmay capture inputs and provide outputs as described herein. More specifically, the sensing arrayand image sensor, described with respect to, and load cell arraydescribed with respect to, may be input devices. Additionally, the output devices may include the display, speaker, and lightsdescribed with respect to. The networkmay be an electronic communication network, such as a local area network, wide area network, and the like, for processing communications with the scale manager. In some examples, the networkmay be wired, wireless (e.g., wi-fi, Bluetooth, or cellular), or some other computer communication network.

1100 1100 In some embodiments, the scale managermay be a server computer or similar device without a user interface but which receives requests from other computer systems having one or more user interfaces. Further, in some embodiments, the scale managermay be a portable computer, laptop, tablet computer, pocket computer, telephone, smart phone, or the like.

An example neonatal care system includes a processor and a memory. The memory includes instructions configured to cause the processor to receive strain amounts corresponding the strain sensors of a tactile sensor array that is disposed beneath a patient that is disposed on a mattress. Additionally, the instructions cause the processor to generate a strain map comprising a visualization of the strain amounts in association with the corresponding X, Y coordinate locations of the strain sensors. Further, the instructions cause the processor to generate a weight map based on the strain map and a strain-weight model. Additionally, the weight map includes weights corresponding to the strain amounts. Further, the instructions cause the processor to identify a patient location. Additionally, the instructions cause the processor to determine a weight of the patient based on the patient location and the weight map.

In one example, the instructions are executable by the processor to determine that the weight map indicates a presence of an additional uniform strain on the strain sensor array, determine an amount of uniform weight corresponding to the uniform strain, and subtract the uniform weight from each of the weights.

In one example, the instructions are executable by the processor to determine that the weight map indicates a presence of an additional gradient strain on the strain sensor array, determine gradient strain increases associated with a sloping subset of the strains, subtract the gradient strain increases from the sloping subset of strains, and generate the weight map based on a strain map resulting from subtracting the gradient strain increases.

In one example, the strain sensors include piezoelectric tactile sensors.

In one example, identifying the patient location includes identifying a contiguous set of strain values that correspond to a predetermined threshold that is indicative of a strain that the patient places on corresponding strain sensors.

In one example, determining that the predetermined threshold is indicative of the strain includes determining a density of the patient, determining a weight value threshold for a sensor based on the density and a number of sensors on which the patient is located, and determining that each of the weights corresponding to the contiguous set of strain values is less than or equal to the weight value threshold.

In one example, determining the weight of the patient includes summing a subset of weights of the weight map that are associated with a subset of strain sensors corresponding to the patient location.

In one example, identifying the patient location includes capturing an image of the patient on the mattress, identifying the patient within the image, overlaying the image on the strain map, and identifying a subset of strain sensors corresponding to where the patient and the subset of strain sensors overlap.

An example method includes receiving weights corresponding to load cells of a load cell array that is disposed beneath a patient that is disposed on a mattress. Additionally, the method includes identifying a patient location. Further, the method includes generating a weight map having the weights. Additionally, the method includes determining a weight of the patient based on the patient location and the weight map.

In one example, the method includes determining that the weight map indicates a presence of an additional uniform weight on the load cell array. Additionally, the method includes determining an amount of the additional uniform weight. Further, the method includes subtracting the uniform weight from the weights of the weight map.

In one example, the method includes determining that the weight map indicates a presence of an additional uniform weight on the load cell array. Additionally, the method includes determining an amount of the additional uniform weight. Further, the method includes subtracting the uniform weight from each of the weights.

In one example, identifying the patient location includes capturing an image of the patient on a mattress. Further, identifying the patient location includes identifying an image location of the patient within the image. Additionally, identifying the patient location includes overlaying the image on the weight map. Further, identifying the patient location includes identifying a subset of the load cells that overlap with the image location of the patient in the overlaid image.

In one example, identifying the patient location includes receiving strain amounts corresponding to strain sensors in a strain sensor array. Additionally, identifying the patient location includes generating a strain map comprising strain amounts associated with corresponding X, Y coordinate locations for the strain sensors. Further, identifying the patient location includes identifying a patient location based on the strain map, and identifying a subset of the load cells that correspond to a subset of the strain sensors that overlap with the patient location.

In one example, identifying the patient location comprises identifying a contiguous set of strain values in the strain map that correspond to a predetermined threshold that is indicative of a strain that the patient places on a corresponding plurality of strain sensors.

In one example, determining the weight of the patient includes summing a subset of the weights associated with a subset of strain sensors corresponding to the patient location.

An example computer-readable storage medium includes instructions executable by a processor to receive strain amounts corresponding to strain sensors in a strain sensor array that is disposed beneath a patient that is disposed on a mattress. Additionally, the instructions are executable by the processor to generate a strain map having strain amounts associated with corresponding X, Y coordinate locations for the strain sensors. Further, the instructions are executable by the processor to generate a weight map based on the strain map and a strain-weight model, the weight map having weights corresponding to the strain amounts. Additionally, the instructions are executable by the processor to identify a patient location. Further, the instructions are executable by the processor to determine a weight of the patient based on the patient location and the weight map.

In one example, the instructions are executable by the processor to determine that the weight map indicates a presence of an additional uniform weight on the strain sensor array, determine an amount of the additional uniform weight, and subtract the additional uniform weight from each of the weights of the weight map.

In one example, the instructions are executable by the processor to determine that the weight map indicates a presence of an additional gradient strain on the strain sensor array, determine gradient weight increases associated with a sloping subset of the weights, and subtract the gradient weight increases from the sloping subset of the weights.

In one example, the strain sensors include piezoelectric tactile sensors.

In one example, identifying the patient location includes identifying a contiguous set of strain values that correspond to a predetermined threshold that is indicative of a strain that the patient places on corresponding strain sensors.

As used herein, the term, mechanism, can encompass hardware, software, firmware, or any suitable combination thereof. In some embodiments, any suitable computer readable media can be used for storing instructions for performing functions and/or processes described herein. For example, in some embodiments, computer readable media can be transitory or non-transitory. For example, non-transitory computer readable media can include media such as magnetic media (such as hard disks, floppy disks, etc.), optical media (such as compact discs, digital video discs, Blu-ray discs, etc.), semiconductor media (such as RAM, Flash memory, electrically programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), etc.), any suitable media that is not fleeting or devoid of any semblance of permanence during transmission, and/or any suitable tangible media. As another example, transitory computer readable media can include signals on networks, in wires, conductors, optical fibers, circuits, or any suitable media that is fleeting and devoid of any semblance of permanence during transmission, and/or any suitable intangible media.

This written description uses examples to disclose the invention, including the best mode, and also to enable any person skilled in the art to make and use the invention. The patentable scope of the invention is defined by the claims, and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.